Loopholes in Google’s data safety labels | Hacker News Reader

Loopholes in Google’s data safety labels

foundation.mozilla.org

191 pointsnemoniac103 comments

> The form includes significant loopholes, like failing to require the apps to report data sharing with “service providers.”

> As we’ve noted, data sharing with “service providers” is exempt from disclosure requirements.

> Many different companies act as service providers, including Google itself

Well, whaddya know. Apps don't have to report whether they share data with Google.

I found it striking that in the list of four questions that Mozilla asked Google, not one of them elicited a straight answer. I don't mean just evasive or slippery; they failed to answer the question, instead answering a different question, one that wasn't asked.

they failed to answer the question, instead answering a different question, one that wasn't asked.

Politicians do this all the time. Constantly. They do it because the PR firms taught them to do this, and those same PR people work for tech companies now.

(It's awful that journalists these days let politicians get away with it. When I went to journalism school, we would get a failing grade if we allowed an interview subject to dodge a question like that, but that's a rant for another day.)

There's a funny passage in one of Al Franken's books about this. Franken was a comedian who ran for the U.S. Senate and was successfully elected. (He later escaped in celestial disgrace, but that's unrelated.) He talks about how he was trained to do this by his PR folks, thought it was stupid, tried it out in a few interviews to get them off his back, and then was amazed that it worked every time.

The documentary "Thank You For Smoking" illustrates this.

There was a funny video just about this not that long ago regarding housing prices in canda:

https://www.youtube.com/watch?v=X5ExYQ21sGY

The sad part is, that the politician asking the question does a better job that most journalists and points at the absurdity of the dodged answers.

Yet it’s all still deeply impersonal. I would have thought it would be much more effective for the one whose questions are being rudely ignored to deliberately make it personal in response: to bluntly impugn the honour and integrity of the other, politely offering grave insult to match the insult received by the ignoring of the question—the sort of thing that would probably have led to a duel in bygone ages. (I wonder if such bodies have any specific rules against referring to someone as “the dishonourable minister” or similar, or anything about means of address in general.)

Mozilla itself is no stranger to weasely language in some privacy-related aspects.

That, or they do fun stuff like giving this splash screen on a new update:

https://imgur.com/a/p4CVBHb "Power and privacy to the people. No need to dig into your security settings. Fierce privacy is our default."

You know, the browser that defaults to Google search and having search suggestions on. I know I'd have a couple privacy settings to change.

> You know, the browser that defaults to Google search and having search suggestions on. I know I'd have a couple privacy settings to change.

And having opt-out telemetry with very limited messaging that it occurs.

To be fair, I think less technical users would think that Firefox was either broken or incapable of basic functionality (leaving a bad impression either way) if search suggestions weren’t on by default. It probably wouldn’t hurt for the default search engine to be something else though, I think most people are used to changing that.

I find search suggestions positively hostile, and is one of the first things I switch off.

I see that Mozilla has started pushing Pocket again. With the last upgrade, I got a new tab with a full-page Pocket ad. They don't need telemetry to discover that I've never used Pocket. I dislike their custom of using upgrades to stuff an ad in my face; and I dislike that scarce screen real-estate is used for an unnecessary button. If they've got to create a new tab on upgrade, it should show the changelist, not a "promotional message".

It's a bloody plug-in, and it should be possible to unplug it.

The fairly clearly implied answers:

A1: we never review the disclosures. We have reviewed zero apps for discrepancies in the last year.

A2: since we never check (refer to A1), it can get as bad as it likes, and would you please be quiet because we’re trying to remain wilfully ignorant over here!

A3: we have taken zero enforcement actions in the last year (surely you understood that from our first answer?).

A4: ARGH! WHY ARE YOU TELLING US ABOUT PROBLEMS? WE DON’T WANT TO KNOW OR WE MIGHT HAVE TO DO SOMETHING ABOUT IT! Y’know. Something.

When it comes to Google, here is another anecdote (loophole?) that I personally experienced once in my life.

I worked for a company whose business Google account was somehow deleted. The access to Gmail was entirely lost, which was super critical for the company. One of our employees spent hours dealing with Google customer support. If I'm not being mistaken here, the response that they kept getting from Google was something along the lines of "due to our data safety and data retention policies, we are not able to recover your Google account". That same phrase was repeated again and again by various Google reps. Long story short, the company recovered everything. The solution was to email one of the investors and ask them if they can help in any way.

It was hilarious. The mail thread was something like:

Company:

> Hey Joe, due to xxxxx, we lost access to yyyyy@yyyyy.com. Google customer support cannot do anything. Any ideas what shall we do?

Joe forwards the email to an internal Google employee with just two words:

> Handle this.

The account is fully back online within 30 minutes.

Curious, what was Joe's relationship to Google? How was he able to exercise that kind of authority?

AFAIK, Joe does not have any direct relationships with Google. I suppose, he's just a powerful venture capitalist.

Who happens to know the random engineer who controls accounts?

Probably a C level. They don’t need control over accounts to get an account recovered.

Glad to see someone trying to keep Google accountable! They're a long way from their Do No Evil mantra these days.

Today's google would be banned by 15 years ago's google. One rule they had was that the style of the ads on your website should not match the style of the website (clearly visible ads). Today most people do not realize that the first page on google.com is mostly ads.

Unfortunately, as is almost always the case, the almighty dollar skews ideals. Is it even technically possible for a public company to be "good" these days?

Logically no, when the overly optimized and less secured options are rewarded.

There's a story of a car company which calculate the cost needed to recall a defective model (safety-wise), versus the fines needed to pay times the possibility that the error may occur. There's cheaper on the later so they don't do the recall.

And the recent train derailments, many people said that if the fines are lower than the reduced cost they get for lowering the safety, they'll keep doing it again.

Is it even technically possible for a public company to be "good" these days?

Yes, there are many companies that do good things, and some that exist solely to do good things.

The difference is that they focus on doing those good things, so they don't show up in headlines very often.

Perhaps: https://www.bcorporation.net/en-us/

There is this fixation on “the dollar” or even “capitalism”, but in reality both of those are essentially just abstractions for the underlying human behaviors, which is mostly greed and the lying and deception used to obfuscate greed.

Neither the dollar as the object of transaction nor the system of accounting (accounting for resource allocation) called capitalism, have any real connection to the underlying drivers. The same problems existed in many other means of transacting (currency) and economic systems (communism notoriously being a sprint towards corruption).

Unfortunately, people have forgotten or have intentionally been misled from realizing the tacit type of self-regulating controls of a system like actual capitalism (you lose money when making bad decisions, you aren’t rewarded by thieving friends in government) and even aristocracy, where you had to maintain a certain balance of relationships even with the peasants in order to prevent being made a foot shorter.

Ironically, the rather tyrannical caste that has we emerged on Mt Davos is forgetting that sometimes the dumb and easily manipulated peasantry the technocrats see themselves as being benevolent demigod leaders of, don’t take so kindly to their shenanigans, with dire consequences for all.

I mean what are you, a bigot, racist, sexism homophobe for questioning the beneficent and glorious, exhaled rule of the Ruling Gods over your life and limitations on what you can do, say, and think? You better. Don’t get any ideas or your gods on Mt Davos will strike you down with ban hammers.

> Neither the dollar as the object of transaction nor the system of accounting (accounting for resource allocation) called capitalism, have any real connection to the underlying drivers. The same problems existed in many other means of transacting (currency) and economic systems (communism notoriously being a sprint towards corruption).

Capitalism provides a moral justification for the actions of greed, it makes responding to those incentives seem inevitable and necessary and so not the responsibility of the person actually doing it.

Neither of those is necessarily inherent to capitalism, I guess, but they are certainly part of the structure of the capitalism we live with so shouldn't be dismissed either.

Capitalism isn’t a form of accounting it’s a mode of production

That is absolutely false and at the core of the rather schizophrenic mentality towards the world today. Capitalism has absolutely nothing to do with the means of production beyond it being a method for accounting for resources and allocating then based on a value assigned, ideally in a kind of double blind manner, which is the part that has been broken by cheating through things like corruption and “money printing”, i.e., fraud.

You may be referring to the favorite alternative method for accounting under the inherently fraudulent system that fraudulently calls itself communism, while hating and having nothing at all to do with community, which allocates resources, i.e., means of production, by central committee that sees itself as some kind of contemporary gods on earth.

The means of production are a separate thing and are generally separate from gore they are accounted for and allocated for some purpose.

To simplify, you having the skill, told, and wood to create a bookshelf are the means of production; your method and decision making that lands you on making a bookshelf instead of book ends or something else, is where you could use capitalist or “communist” practices to determine that, Party of which is accounting for the resources and drawing a conclusion as to what to do.

That's why regulation is needed. Efficient corporations will (and must) do the most profitable and legal thing always. What would the economy look like if corporations behaved irrationally, by doing unprofitable things based on unpredictable/fickle human emotions? It might be a cynical way to look at it, but overall I don't think it's sustainable to place the responsibility of caring for society on corporations.

Saying that Google or their employees are evil for doing the thing they're supposed to do is pointless. If they don't do <evil thing>, a competitor will do <evil thing>. The only way to save the world from <evil thing> is to make it illegal (and preferably with criminal penalties, not just small fines), so that corporations can avoid doing it without worrying about the competitive risks.

(although Google and all of big tech are evil monopolists and they need to be broken up)

Governmental regulatory agencies are largely captured by the likes of Google and Apple, and if not completely captured there's a very large influence that prevents regulators from taking any actions that would actually threaten profits of monopolistic corporations. There has been a little push back recently, in terms of preventing more mergers and a push towards breaking up conglomerates, but not without a lot of political resistance:

https://mattstoller.substack.com/p/on-lina-khan-derangement-...

> It might be a cynical way to look at it, but overall I don't think it's sustainable to place the responsibility of caring for society on corporations.

This line of thinking bothers me more than it probably should. How can corporations be thought of as separate entities from the people it consists of, in any context except legal? I wrote a lot more, but decided to distill it into one question: How can they be, in good conscience, allowed to get a pass?

Because corporations aren’t people, they’re machines that happen to be powered by human labor. There’s some wiggle room for individuals to make a difference, but ultimately the market will win in the long run.

Consider the CEO of Google growing a conscience (hypothetically ofc). If he decides to cut back on <evil things> for the good of society, then society will benefit, but Google shareholders will lose. The impact will be measurable in financial statements, and when the board meets to vote on a new CEO, they’ll replace him with someone who will do a better job of making those numbers go up (aka someone without that conscience). The system self-corrects in the long run!

The CEO has a legal obligation to protect the interests of share holders (“fiduciary duty”). So from that perspective, a CEO acting in the interests of society rather than shareholders is breaking the law, and could be personally liable.

The incentives in the system are not set up in such a way that corporations could be reasonably expected to “self police” in a way that benefits society at the expense of shareholders. Sure, in some cases it’s possible to have a win-win scenario, where everyone benefits, but those are extremely rare.

IMO this is a good system because it leads to the primary producers to perform at peak efficiency, but it doesn’t work unless it’s balanced by government regulation, which is something that the government has been miserably failing at for the past two decades (at least)

Going back to the question you're responding to:

> How can [corporations] be, in good conscience, allowed to get a pass?

I sort of expected you to answer that corporations don't get a pass. I agree with your premises:

1) corporations will follow their own incentives

2) legislation is the means for "caring for society"

3) legislation changes a company's incentives and should be enforced

It goes against the Zeitgeist simply because Americans understand that their representatives and senators are largely corrupt (regulatory capture) so enforcement and legislative remedies are a pipe-dream until we deal with legislative corruption.

This is the central problem with investment capitalism, as distinguished from what I'd call industrial capitalism. Once you put the shareholder associations in the driver's seat, and eliminate most input from the people who work for the corporation, you always find the company trending towards monopolistic behavior, eventually leading to poor-quality products with various safety issues, because that's a way of reducing costs once you have a captive market. With Google and Apple, ownership is somewhat diffuse but the top tiers are the usual suspects: Vanguard, Blackrock, FMR, State Street, Berkshire-Hathaway, etc.

https://finance.yahoo.com/quote/GOOGL/holders/

https://finance.yahoo.com/quote/AAPL/holders/

The only real solution is to give the employees of the corporation the degree of power when it comes to selecting the executive cohort as the shareholders of the corporation have - i.e. a 1 : 1 ratio. Germany does this to some extent.

The other option is more radical - eliminate the shareholder system of devolved liability entirely, and only allow the people who work within a corporation to own shares in that corporation. (Note that the traditional communist-socialist approach would be to simply replace the shareholders with a governmental central committee while keeping the structure essentially the same.)

> Today most people do not realize that the first page on google.com is mostly ads.

Ugh this. Trying to search for almost any generic phrase or term just floods the results with links to shops and "TOP 10 BEST thingYouSearched"

Google Search feels almost useless now for anything other than shortcuts [in]to websites you already know about.

Google feels like an index engine of Reddit (and sometimes HN) to me.

> Google Search feels almost useless now for anything other than shortcuts [in]to websites you already know about.

In my experience, not even that. I used to be lazy and use google search to get to websites I knew, but it can no longer find them even if I type it in exactly! So I've mostly had to revert to making sure I bookmark the sites and go in that way. (And of course I've switched to ddg for search.)

Trying to search for almost any generic phrase or term just floods the results with links to shops and "TOP 10 BEST thingYouSearched"

eBay is the worst for this.

"Pancreatic cancer now on sale at eBay!"

"The best Libertarian Morals now on sale at eBay!"

"Buy discount Rings of Saturn now at eBay!"

I would gladly pay for one of these.

It was "Don't Be Evil".

Take away facts:

    Google's business model is built around privacy invasion and data collection.

    Mozilla's business model is built around Google.

Process these as you see fit.

You're not entirely wrong, but how do you reconcile that with the OP? Genuine question.

In the same way you reconcile "Don't Be Evil" with Google.

In the same way you reconcile Mozilla's privacy concerns with the fact that every Firefox install gets tagged with a unique digital fingerprint and keeps a persistent websocket connection open in the background.

> In the same way you reconcile "Don't Be Evil" with Google.

I don't think that's fair, especially since they have already dropped it.

> In the same way you reconcile Mozilla's privacy concerns with the fact that every Firefox install gets tagged with a unique digital fingerprint

I'm not sure what you're on about here. Care to elaborate?

> and keeps a persistent websocket connection open in the background.

Hadn't heard of this one before. Do you have sources?

I don't think that's fair, especially since they have already dropped it.

Dropping it doesn't change the duplicity that existed for almost 2 decades.

Hadn't heard of this one before. Do you have sources?

https://borncity.com/win/2022/03/22/firefox-installer-weist-...

https://www.scss.tcd.ie/Doug.Leith/pubs/browser_privacy.pdf

> Dropping it doesn't change the duplicity that existed for almost 2 decades.

It was dropped in 2015, 17 years after Google's foundation. Are you saying they were evil from the beginning? I couldn't tell but at least for their first decade or so they seemed to be trying to do good. In any case, it's very different to be Google and evil vs. Mozilla/Firefox having a unique ID in some downloads.

Why do I say some downloads? Because that ID is assigned when you click the link on their website's downloads page. If you get it from your package manager, which I suspect a very large portion of their user base does, you don't get uniquely identified. If you download from any mirror or otherwise free software download store, you don't get uniquely identified.

If you download from any mirror or otherwise free software download store, you don't get uniquely identified.

What you are downloading in most cases is just an installer executable -- no software included. The installer pulls the actual software from Mozilla's server. Each copied served is given a unique ID and is unlike any other -- easily proven by hash or checksum (Mozilla calls this a "dltoken", see my reference above).

I've just sampled some mirror sites.

- https://mozilla-firefox.fileplanet.com/ - Indeed it downloads a download manager.

- https://download.cnet.com/Mozilla-Firefox/3000-2356_4-102085... - Downloads an installer. It's for an older version but shouldn't have the unique ID, and it should be able to update itself but I haven't tested.

- https://techviral.net/firefox-offline-installer/ - Also the installer, and for the latest version.

That's 2 out of 3 without an unique ID.

You have also conveniently glossed over the package manager angle, which I suspect is a pretty large slice of their user base.

Google replaced "don't be evil" with "do the right thing", which is entirely fair to hold them to.

Firefox install fingerprinting: https://www.ghacks.net/2022/03/17/each-firefox-download-has-...

Persistent web socket connections: https://www.scss.tcd.ie/Doug.Leith/pubs/browser_privacy.pdf (page 8-9)

In both cases this "telemetry" is something that the user did not ask for or authorize, nor is it disclosed prominently.

> Google replaced "don't be evil" with "do the right thing", which is entirely fair to hold them to.

That's fair, but "the right thing" for whom? If it's for their shareholders, which apparently by law is in fact not only the right thing but a requirement, then it's hard to argue they're not good on their word.

> Firefox install fingerprinting: https://www.ghacks.net/2022/03/17/each-firefox-download-has-...

I commented on the sibling. This is only true if you download it directly from their website. Firefox is acquired from many other sources - package managers, FTP sites, other free software distributors...

> Persistent web socket connections

Thanks, this does seem less than ideal. There seems to be zero evidence of any breach of anonimity via this mechanism though. Just to be clear, I agree that it shouldn't be this way.

> In both cases this "telemetry" is something that the user did not ask for or authorize, nor is it disclosed prominently.

No argument here. I both agree with your assessment and that this is a bad thing.

In a space without good regulation I guess this is as much as the free market can deal with, It's no solution though, all problems are nails in this

Arguably, Mozilla's business model is built around getting people to switch to Firefox so they have leverage to make Google pay more for being the default search engine.

Arguably, this is an inadequate description of reality.

Firefox's market share has done nothing but decline for decades --- and Google still pays.

Unless Google pays more for the reduced market share, your argument doesn't stand.

The fact that Mozilla are failing at their business model doesn't mean that's not the model.

Unless Google pays more for the reduced market share, your argument doesn't stand.

So my argument stands.

Mozilla's current deal with Google is good through 2023 and is worth around $400-$450 million per year.

Every deal Mozilla has made with Google has payments increased --- even as Firefox's browser share has decreased.

https://www.androidheadlines.com/2020/08/mozilla-firefox-goo...

I haven't seen numbers that back that evolution over the years.

This article [0] from 2018 suggests an uptick, but it justifies it as "a deal to get paid for Google search traffic in parts of Europe" - so, actual market expansion.

I'm not saying you're wrong, only that I haven't seen evidence that you're not.

[0] https://www.cnet.com/tech/services-and-software/google-firef...

Google pays for Firefox's livelihood to prevent anti-trust coming for Google Chrome.

Google does not care about Firefox's market share. Google only cares that there is an appearance of competition in the browser market.

Then why do they pay Apple for the default search on Safari? Or shy did Yahoo buy the default search on Firefox for several years? And why are Google focusing on a Chrome monopoly when buying the default has out them at risk of being found a search engine monopoly?

The market has grown, I wouldn't be surprised if their amount of users has stayed similar for quite a few years now. Google pays per user, not per percent of market share.

Firefox's total MAU is still dropping over time.

This is their own data on the matter... https://data.firefox.com/dashboard/user-activity

They've lost about 50 million users between 2019 and 2023 - roughly 20% of the 2019 totals.

In my opinion - Google pays because having a well known name that they can claim is still competing in the browser space is beneficial to them with regards to antitrust and monopoly laws. The ad revenue from the Firefox user base helps to make that cost manageable.

Google's business model is absolutely not built around privacy invasion, and their model is built around data collection of public data. If, for some reason, they couldn't use personalized ads anymore, they would simply earn less, but their business model would be the exact same as before.

Mozilla business model is entirely built around their market share. As such, they are in a much more difficult situation than Google's Chrome team. If Mozilla has a difficult time (financially), they are more likely to do shady things than the Chrome team which has no such pressure. For example, the Chrome team would never accept any cryptocurrency deal; something that I have no difficulties imaging Mozilla to do.

Obviously, Chrome will not actively hurt Google's other businesses (specifically ads), but Chrome's primary goal is to make the web a better place. The worst case scenario for Google is that the web (and thus their revenue stream) disappears.

Ah yes, the old "to make the world/web a better place".

For the sake of children, to defeat the terrorists... :)

My recent heuristic is that any short comment starting "Ah yes" is a strong signal of low quality or trolling argument.

What do you reckon: https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...

> The form includes significant loopholes, like failing to require the apps to report data sharing with “service providers.”

Mozilla is taking a principled stance that _all_ sharing of data should fall into the same category of "we gave your data to someone else", but I think there's a much more pragmatic breakdown. Data can either be not-shared, shared to service providers with contractual obligations to treat that data as owned by the original service, or shared to service providers who may then use and further share that data for their own purposes.

Examples of the latter two would be the difference between sending crash reports to a crash reporting service that doesn't interpret user data in any way and is only a mechanism used by the original service, vs, say, sending it to Facebook to report an advertisement conversion, where it will be used to further the Facebook ad modelling.

As a privacy conscious user I typically don't mind the former, but despise the latter. Mozilla does not appear to distinguish between these two cases. One could argue that the former is still increasing the scope of data being hacked, or that companies might breach their contracts and use it anyway, but if you assume illegality then most of this breaks down anyway, we have to assume that companies stick to the law and their contracts for any of this to be of any use (and I think that's mostly reasonable for us to do).

I work closely to this area at Google so I don't want to comment on how this fits with Google's data safety labels and the privacy policies of apps. I think people should read the documentation and draw their own conclusions.

I personally think that this issue sits at the intersection of principles and pragmatism – making information correct but conveying a less immediately useful message, vs making information accessible and actionable for the majority of users. Google (and Apple) also still need to appeal to developers to get them to build for their platforms, and that again may require pragmatism.

> Data can either be not-shared, shared to service providers with contractual obligations [...], or shared to service providers who may then use [...].

To follow up, this distinction is also made by both GDPR in Europe, and CCPA in California. (In fact, CCPA uses the same term "Service Provider" and I wouldn't be surprised if that's where Google got it from.)

Still, both GDPR and CCPA require disclosure of Processors/Service Providers.

Transfers to a Processor/Service Provider have restrictions, and that's significant. Such third parties are treated differently under the law, with those restrictions removing the need for other safeguards. But the don't entirely remove any risk.

Ultimately one of the goals of these disclosures is letting the user take control over how their data is used and distributed. Some users (including you, and including myself) are comfortable with data transfers that are protected by legal safeguards rather than technical ones. Some are not comfortable with that, and have valid reasons to be. Some users reasonably want to know their full security risk exposure, which means being aware of which parties are handling their data. Some users simply don't want to do business with certain platform providers, even indirectly.

I'll also point out that the EU has found that US safeguards about transfers to Service Providers are inadequate, because if a third party has your data that means US law enforcement can get your data from that third party. So there are very valid concerns that the safeguards Google considers adequate for a Service Provider may not satisfy all users.

All of which is to say, I agree that the distinction about Service Providers is useful and meaningful, but I think it is the wrong decision to exempt Service Providers from being listed altogether.

Thanks, this is a good breakdown.

It sounds like, apart from companies breaking the law, the only caveat to service providers being "safe" from user data sharing is law enforcement, and in particular US law enforcement.

While this is something I care a lot about, I think many people would consider data use for profit, and data use for law enforcement substantially different, so it _may_ be reasonable for Google not to count law enforcement in their system design. This is a hypothetical, I don't know if this was a factor and I don't really hold this opinion, but I could understand it.

> As a privacy conscious user I typically don't mind the former, but despise the latter. Mozilla does not appear to distinguish between these two cases. One could argue that the former is still increasing the scope of data being hacked, or that companies might breach their contracts and use it anyway, but if you assume illegality then most of this breaks down anyway

You have to assume not necessarily blatant illegal behavior, but instead a slow stretching of the law. There is huge financial motivation for a company to switch a data pipeline from the former use case to the latter. Those incentives will push them towards changing the fine print or creative interpretations of that fine print, or will get them acquired by a more profitable, less scrupulous company willing to do so. Look at the incentives, and that will tell you what the outcomes will eventually be.

The only way to avoid this is either not sharing the data in the first place, or sharing the data only after it's encrypted so that the service provider cannot help but treat it as owned by the original service. Anything else is fighting a losing battle.

I get that these things can change over time, and there are incentives to do so, but this is why we must take legislation like the GDPR seriously, companies must do their due-diligence on contracts with service providers, and we should hold those breaking contracts and/or the law to account.

I don't "people are going to break the law so let's not do anything" is the most pragmatic response. I don't disagree, and can certainly empathise with it, but I do see it as closer to Mozilla's principled stance than to what I think most people would be happy with (assuming they had the facts, which most don't).

> Our research closely mirrors the 2021 Washington Post investigation into the accuracy of the Apple App Store’s new data privacy labels for apps. The investigation found that not only were many of the labels confusing and incomplete - they were also false

Apple came up with this "nutrition label" idea for app usages of data that is self reported by developers. Everyone assumed "increase in privacy". Google copied the practice and it's full of "loopholes" now.

This whole data privacy, unlike cryptography, has become more of a marketing tactic and media loves to create super villains and super-heros out of it. There's no substance here.

Is there really a double standard here? The subhead of the wapo article is:

> Apple’s plan to make iPhone apps be transparent about the data they take falls short of being helpful — or even accurate

Funny coincidence that I have received a notification from the Play store this morning saying that analysis of my app (with super tiny outreach) showed that my Data Safety Form was inaccurate and asking me to fix it asap.

Seems like letting the user enter an email address to recover his password counts as email collection (he cannot subscribe, provide or see his email using the app, hence the initial form I submitted).

If you're sending the email address to a server (which is presumably needed to recover a password), I think this makes sense. Apple/Google has no way of knowing or verifying what the the server is doing with that data.

Absolutely, I meant that the app is not used to collect the email address since it is already in our system but it is true they have no way of knowing.

Google's and Apple's central rationale for limiting app developers to working through their app stores is that they're screening apps for security reasons. If they're doing such a poor job of it, then there's no reason for such limitations, other than as another rent-seeking monopoly effort.

> "Additionally, a new Commerce Department report cited Apple and Google as “gatekeepers” of mobile app stores, leaving consumers at the mercy of their pricing and selection. The report calls for new legislation that would boost competition in the mobile app market, which would help both consumers and app developers."

To extend the FDA analogy, we don't really require stores that sell products to be held responsible for flaws in or problems with those products, e.g. contaminated baby formula. The onus should be on the manufacturer (or the app developer) - but this also means that if a store sells a certain brand (e.g. Whole Foods 365 brand) then they are directly responsible for any issues with that product.

It is the same situation with Apple's App Store privacy labels.

shush...

While privacy policies are increasingly tailored to an app's actual data usage,* there are still FAR too many out there full of the same copypasta allowing a company to do whatever they want with your data. With that in mind, it's hardly a surprise that there's a disconnect between what engineers self-report and what's buried in the policies.

* Based on my own experience over the years actually reading them as someone who both cares about this stuff and works in the space.

I don't have specific views on what they've written.. but 90% of donations to Firefox come from Google. The 3 year contract ends in 2023..

If Google decides to not fund it next year it's bad management on FFs part.

Putting all your eggs in one basket is very risky and puts you in a place where you may have an ugly compromises Vs face bad consequences..

> 90% of donations to Firefox come from Google.

Why? What does Google get from it? How do they justify this expense internally?

Google is the default search engine for official FF binaries.

Oh, but that wouldn't be just a "donation" then..

They pay to be the default search engine. It's just the cherry on top, that as long as Firefox survives, Google can deflect browser-monopoly accusations.

Google is currently facing a search engine monopoly case where their purchase of the default search in other browsers is being used against them.

Being the default search has value. It's why Google pays Apple billions a year for the same thing.

I would think the monopoly deflection is much more valuable to Google than being Firefox's default search engine. 3% marketshare, comprised primarily of, if I had to guess, people who are the most likely userbase of any browser to know how to change their default search engine.

For one side it's an existential question, the other side its just a rounding error..

There is no explicit loss of value for Google, they save money + get more chrome users..

Now if MS comes into play or are already playing in the background, then is a totally different story..

> we found discrepancies between the apps’ privacy policies and the information they reported to consumers on Google’s Data Safety form.

So its like a "sue yourself" system? :-D

Consumers don’t read privacy policies so they shouldn’t be a major part of this. Something more akin to the nutrition label example they use is needed.

Somewhat related to the original post, our research team at Carnegie Mellon University has been investigating how well developers can fill out these kinds of privacy nutrition labels.

This paper looks at iOS developers. We found that there was a lot of misconceptions and misunderstandings that led to under-reporting as well as over-reporting of behaviors. For example, on iOS, "tracking" is defined only in the context of advertising. A lot of people also misunderstood what "Data Linked to Users" means https://dl.acm.org/doi/fullHtml/10.1145/3491102.3502012

Our research team has also been crawling these labels on both Apple App Store and Google Play to understand changes in labels over time. One early result is that these nutrition labels are not updated very often. https://dl.acm.org/doi/fullHtml/10.1145/3491101.3519739

Lastly, we've developed an IDE plugin for JetBrains to help developers fill out labels more accurately (and hopefully more easily too) for Google Play. The key idea is to have developers add privacy annotations (in the form of Java annotations) directly into their code. The plugin uses a bunch of heuristics to flag where you might need an annotation, and helps you fill out those annotations. These annotations are then used to generate a CSV file that can be uploaded to Google Play. https://matcha-ide.github.io/

I also have a talk called Helping Developers with Privacy. It summarizes the last decade of our team's research on this topic. (Note that Slideshare has become a bit spammy, inserting ads in between slides, sorry about that) https://www.slideshare.net/jas0nh0ng/helping-developers-with...

Stepping back, our team has been looking at two big ideas for privacy. One is these privacy annotations. I think annotations has the potential to be transformative for the privacy ecosystem. They can help developers, they can be assisted with IDE tools, they can be embedded into compiled code, they can be used to help auto-generate privacy user interfaces, and can also be used as a major hint to check the behavior of the apps (by app stores, by auditors, by journalists, and by researchers).

The second is that every app should have a simple manifest that whitelists an app's privacy-related behaviors. The manifest describes what data will be used and a chain of transformations, similar to Unix pipes. In the context of smart homes, an example might be: get access to microphone -> transform to loudness -> send to sleep.com. While we don't know exactly what sleep.com will do with the data, we can have more assurance that it's not sending raw microphone data. Here's a paper at Oakland Security that describes our ideas. An interesting part of our initial analysis is that the vast majority of apps don't need raw data (raw microphone data, GPS data, logs, etc). Rather, they need a processed and more granular form of it (e.g. loudness, what city you are in, etc). http://haojianj.in/resource/pdf/peekaboo-preprint.pdf

Happy to chat more about these topics, we've been working on privacy for quite a while and have had some successes with industry and with the US Federal Trade Commission.

yeah this stuff is meaningless until it's asserted from a linter or runtime firewall

I'm not sure why mozilla is wasting their time criticizing detail mismatches -- the self-report system only creates accountability for relatively well meaning players who are large enough to 1) be worth auditing and 2) have reputational risk

little snitch everywhere

Perhaps if they had focused on developing Firefox OS they wouldn't have to put up this.

Well, they don't mind selling their users to Google or Pocket, so are they really loopholes? or is that a PR stunt? i'm leaning towards the later, what about you?

In what way is Mozilla selling their users to Pocket? Do you realize that Mozilla acquired Pocket years ago?

Mozilla didn't acquire it, Mozilla is a non profit

Mozilla Corporation did, and they put it in Firefox and you can't opt-out

> you can't opt-out

You sort of can, but you need to manually dig through tens of hidden options in about:config.

You can remove the button from the toolbar and disable pocket stuff on the new tab page. What other touchpoints are there that average users should care about and that can only be disabled in about:config?

It's just one option in about:config https://support.mozilla.org/en-US/kb/disable-or-re-enable-po...

The Mozilla Foundation owns the Mozilla Corporation, so no matter how you slice it, yes, Mozilla acquired Pocket.

Anyone else irritated by the "Executive Summary" heading for what I think most would call an abstract? As a non-native speaker I had to look up the term to confirm it meant what I guessed it would mean: summary meant for executives as in a corporation's "executive officers".

Always breaks my heart to see corporate speak creep into normal speak :(

Btw, "Executive Summary" being listed under "Summary" also seems weird. I'd call that a table of contents.

Feeling particularly nitpicky today.

"Executive Summary" is an inoffensive way to say "Look, I know you cannot understand this and/or will not read this, so here's a small amount of text written simply that you can use to make a decision."

Yeah... That's called a "summary". But people made trouble by calling the table of contents a summary, and the name is now ambiguous.

Not exactly. Say for example you’re in a situation where an executive has to approve the budget of an engineering project they are not able to understand. A summary would do no good, because it would be incomprehensible to them.

In other words, an executive summary is the summary for a non-expert stakeholder. In practice you use this to tell the executive what you want them to think.

> In other words, an executive summary is the summary for a non-expert stakeholder. In practice you use this to tell the executive what you want them to think.

...no, an executive summary is not "a summary for a clueless executive". An executive summary provides enough information to act on the material... that is, a summary that promotes executive function[0], otherwise known as decision-making based on what it is summarizing.

The "executive" here is not a noun, it's the same "executive" as in "executive function" and "executive decision". And yes, it's also where the job role gets its name.

[0]: https://en.wikipedia.org/wiki/Executive_functions

> The "executive" here is not a noun

I understand, but in this case that’s a distinction without a difference. For whom is the executive summary? For the person making the executive decision, ie the executive.

In my career, executive summaries are crafted to guide the executive to the outcome desired by the author. Perhaps that’s unique to my experience.

It's just that "the executive" in this case doesn't refer to "someone with the job role of executive", it refers to any reader capable of making their own informed decisions. The role of an executive summary is to inform them enough for them to make a mostly-accurate decision (compared to the decision they would've made if they instead read everything that was being summarized).

Reply on news.ycombinator.com